Autism spectrum disorder (ASD) is a condition that impacts a person’s emotional, cognitive, social, and physical well‐being. Symptoms include challenges in communicating, struggles with social interactions, fixation, and repetitive actions. It is crucial to detect ASD in young children to minimize the impact of the disorder through various therapies focused on behavior, education, and family. The application of artificial intelligence has been important in detecting ASD in children. Previous studies have proposed different methods for identifying ASD, mainly using either demographic information or visual characteristics separately, without effectively combining both approaches. Our study presents a new approach to detecting ASD that takes into account both demographic and visual information. Therefore, a framework was suggested to assess different deep learning models for the early identification of ASD. The proposed framework consists of four modules such as stacked bidirectional long short‐term memory (SBiLSTM) using attention mechanism for representing text/numerical features, multilevel 2D‐convolutional neural network–gated recurrent units (ABM‐2D‐CNN–GRUs) using attention mechanism for extracting facial features, and multimodal factorized bilinear (MFB) pooling for combining the features. Moreover, the conditional probability approach calculates a distinct weight for each class based on specific features, leading to enhanced system performance. In conclusion, the AlexNet CNN has been proposed for prediction and its performance was assessed using the multiactivation function (MAF) framework. In this study, we examined the dataset for screening ASD and the dataset for children with autism. It is crucial to detect ASD at an early stage. We have identified features that can differentiate children with ASD from those without ASD. The suggested system achieves a higher accuracy rate of 99.2% compared to current systems. This outcome indicates that our system is better at predicting ASD compared to other advanced methods.

ALS (Amyotrophic Lateral Sclerosis) is one of the fatal diseases across the world. Therefore, early detection can save patients suffering from ALS from life-threatening consequences. Typically, ALS can be identified based on different factors, and one such factor is voice analysis. Detection of ALS using sound signals is convenient and simpler than other methods, as it is a non-invasive approach, which makes the process faster and more efficient for detection. However, detection of ALS using traditional approaches is challenging, as it is a time-consuming process and heavy reliance on medical experts is needed. Therefore, AI-based models can be used for effective classification of ALS and non-ALS patients, as AI-based models possess the immense ability to examine vast amounts of data, including audio files, effectively. Owing to these factors, the proposed model focuses on employing an AI-based model for ALS classification based on vowel phonation /a/ and /i/. The process is carried out using the Minsk2020 dataset, where important features needed for the proposed model are extracted using MFCC (Mel-frequency cepstral coefficients) by removing the shakiness and jitteriness of the voice. The MFCC feature extraction technique extracts features based on the mel scale, as this reflects human auditory perception, thereby extracting features that are useful for classification. These extracted features are fed to CNN-LSTM (Convolutional Neural Network – Long Short Term Memory) with rapid dilatenet for classifying ALS and non-ALS patients accurately by identifying even the subtle changes in audio signals using maximizing the expansion/dilation rate and aid the context information for interpreting and analyzing the sound of vowels accurately and correctly without any loss of information. Finally, the efficacy of the proposed model is assessed using evaluation metrics. The proposed research work can assist medical professionals in detecting patients with ALS based on vowel phonation.